Research And Application Of Classification Based On Classification Frenquent Pattern Tree

With the rapidly development of information industrialization, data mining plays more and more important roles in many trades. Of each branches of data mining, association and classification are two very active fields with broad application. And there are some comparability between association and classification to find strongly relative item sets. So a new classification method—classification based on association was proposed which use association technology to accomplish classification. And a new algorithm classification based on CFP-tree (classification frequent pattern tree) are advanced in this paper.Firstly, this paper explores main theory and algorithms of classification and association. Secondly, popular classification based on association algorithms CBA and CMAR are introduced. Then this paper put forwards and discuses the new algorithm--the classification based on CFP-tree.In the new algorithm the CFP-tree is formed by adding label attributes to the FP-tree. The CFP-tree is used storing the frequent classification information in the database. Then the restricted sub-tree is built by scanning the CFP-tree from bottom to up according each label attribute. There are not links between the character attribute nodes with same attribute value in the CFP-tree. And the frequent classification information is found from each restricted sub-tree, thus the amounts of the pointers can be largely reduced and the efficient of the algorithm can be improved. At last, the paper gives an instance in the education system by applying the new algorithm to Intelligence for Undergraduate.